Guidelines for the Analysis of Data from the Potentiostatic Intermittent Titration Technique on Battery Electrodes

Abstract

The Li-ion battery (LIB) modeling community needs to rely on a vast bank of measured Li-ion cell parameters. Most of them are determined by routine experiments but some can solely be estimated by fitting a model to a well-designed experiment. For instance, the determination of the lithium chemical diffusion coefficient has been the focus of numerous studies. Specific techniques have been proposed for its estimation, e.g., the potentiostatic intermittent titration technique (PITT), the galvanostatic intermittent titration technique (GITT), the cyclic voltammetry (CV), and the electrochemical impedance spectroscopy (EIS). Our study focuses on the PITT so as to provide guidelines for experimental measurements and corresponding data analysis. The validity of underlying assumptions of published analytic solutions used to fit PITT data is assessed using a pseudo-2D (P2D) model. Among the tested assumptions are the drop of the porous-electrode effect, of the particle size distribution and of the finite kinetics of Li insertion/de-insertion. Tests are also conducted to assess the error made on the chemical diffusion coefficient and reaction-rate constant determination by fitting 2-electrode simulated data with a 3-electrode P2D model. An example of the P2D model fitting of PITT data obtained on a thin graphite electrode is provided.